I’ve chosen Look Me In The Eye to read next. It’s by John Elder Robison, the brother of Augusten Burroughs. Robison has Asperger’s syndrome, a less severe version of autism. It’s a memoir. I’m very excited.
This will be my last “regular” book for a couple weeks because I have a couple computer books en route about MySQL databases. I’ll write more about that in the coming weeks, my recent fascination with data storage and retrieval and all the neat little things I’ve been using databases for lately, ranging from baseball statistics to tracking my daughters cellphone usage. Good stuff. And she’s grounded.
This past week, since I haven’t been reading, I’ve been thinking a lot about reading. I’ve been looking at a few different aspects of my reading so far. Something that I track, but don’t show on the website, are my ratings. I rate each and every book on a 0-10 scale. I also track the number of pages and a rough estimation of word count based on a sample page.
But I started getting curious about certain statistics, mainly if there was any correlation between the size of the book and the rating it received. Would it be more likely that a smaller book would get a better rating because I could read it faster, retain more information, and the idea that smaller books might be more focused on an individual subject, and thus was chosen more carefully? Or would larger books get higher ratings? After all, I’m not going to tackle a longer book unless I know for sure that I’m going to enjoy it throughout, and give it a more analytical inspection before I purchase. Or, because it’s a longer book, I dive into it with more dedication, knowing it will require more commitment than previous books, and in turn, have an emotional buy-in before I even begin reading. I couldn’t see myself spending 12 hours during a given week on a certain book, finish it, and then say, “that was a piece of crap.”
So I decided to take my ratings and the page count, and stick it into a formula called Pearson’s Correlation Coefficient. This formula can measure the correlation between two data sets and tell you how much correlation there is. I had a small sample size, only 25 books, but I plugged in the numbers anyway. The “score” ranges from -1.0 to 1.0, with zero being no correlation and as you move to either side of zero, you get closer to having a correlation (that was very crude but I’m not a mathematician). A -1.0 score would mean as the page count decreases, the rating increases, and a +1.0 would mean as page count increases, rating increases. Everything in between shows a smaller and smaller correlation between the two sets of numbers. If you were to plot these numbers on a graph, you would see the points forming a shape going upwards, left to right. The opposite for a negative correlation.
So I’ve described, roughly, correlation (to the best of my ability). All this only to tell you that there was no correlation at all between the length of the books and the ratings I’ve given them. My Pearson’s score was 0.27 which is no correlation at all. And on a graph it looks like a totally random placement of data points creating no shape at all. X is my rating, and Y is page count.
Now I’m going to learn more about Asperger’s syndrome. Laters.
tip: if this wasn’t fun to read, you’re not alone, but reread it anyways while listening to Flaming Lips’ Free Radicals. it will be much better.